PROJECT SUMMARY This project is responsive to NOT-AG-18-039 Alzheimer's-focused administrative supplements for NIH grants that are not focused on Alzheimer's disease. Nursing homes are an important site of care for persons with Alzheimer?s disease and related dementias. Nearly 1.5 million Americans live in the nation?s nursing homes, with an estimated 11% of residents affected by Alzheimer?s disease. Of Americans with Alzheimer's disease and related dementias, three quarters will spend time in nursing homes before 80 years of age and most will die there. People with Alzheimer?s disease spend an average of two years in a nursing home. A 2011 Institute of Medicine report acknowledges that people with cognitive impairments ?may be unable to convey information to clinicians about their pain?. The parent grant posits that the federally-mandated Minimum Data Set (MDS) 3.0, which captures longitudinal information on virtually all residents in all US nursing homes, can be leveraged to better characterize pain symptoms. We apply advanced statistical methods using a ?resident-centered? approach to identify subgroups of residents with similar pain symptom clusters. The overarching goal of the administrative supplement is to further our understanding of factors that affect quality of life for nursing home residents with Alzheimer?s disease and other related dementias. The proposed administrative supplement seeks to expand the scope of the parent grant to: 1) understand the extent to which pain symptom clusters are qualitatively different for nursing home residents with and without Alzheimer?s disease and other related dementias; 2) develop novel scales using MDS data to capture and describe differences in interactions within the nursing home (e.g., social interactions and social isolation); and 3) evaluate the extent to which measures of social interaction and isolation are associated with pain symptom clusters. The proposed work leverages a national data resource developed under the parent R01 which includes MDS data, Medicare eligibility, and Medicare Part A and D claims from 2011-2016. We will use latent variable methods (latent class models, factor analysis) and hierarchical models to: 1) Estimate the extent to which pain symptom clusters differ for residents with Alzheimer?s disease and related dementias relative to cognitively intact residents while adjusting for potentially confounding factors; 2) Develop novel MDS 3.0 based measures to capture social interactions and social isolation in nursing homes; 3) Evaluate the extent to which differences in pain symptom clusters by Alzheimer?s disease status can be attenuated by social interactions or exacerbated by social isolation in nursing homes. This research represents a novel area of inquiry, and if successful will provide new measures and valuable information to inform quality improvement interventions for improving pain recognition and management in nursing home residents with Alzheimer?s disease or other related dementias.